Association mapping is an approach used to isolate genes associated with diseases, or other phenotypic traits of interest. This method makes use of the statistical associations (linkage disequilibrium) that are often present between linked loci. The strength of these associations, and the recombinational distance over which they occur, is determined in large part by population history: for example, patterns of population growth, or subdivision. For this reason, the success of association mapping depends heavily on population history, however the nature of this dependence is currently not well understood. The proposed research has two interconnected themes. (1) There will be a systematic study of the dependence of linkage disequilibrium on population history, in order to identify optimal mapping strategies for particular kinds of histories. (2) Since detailed demographic information is not known a priori for most populations, a major goal of this research plan is to develop statistical methods for estimating demographic history on the basis of genetic haplotype data. The statistical methods developed for inferring population history will then be used to identify populations which are suitable for mapping studies, and to suggest for particular populations which mapping strategies are most appropriate.